| 
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • You already know Dokkio is an AI-powered assistant to organize & manage your digital files & messages. Very soon, Dokkio will support Outlook as well as One Drive. Check it out today!

View
 

Meeting (2019-01-10)

Page history last edited by Alan Liu 5 years, 2 months ago

 

Meeting Time:       Tuesday, January 10, 2019, 10am -12 noon (Pacific)

Meeting Location: DAHC (Digital Arts & Humanities Commons) (directions)

Meeting Zoom:     We'll use Alan's "instant" Zoom ID (our default meeting Zoom):  https://ucsb.zoom.us/j/760-021-1662

 


 

 

 

Purpose of today's meeting

  • Project roadmap for rest of academic year
  • Review and push forward with preparatory work for next cycle of interpretation workshops 
  • Optional at end of meeting: breakout meetings 
  •  

 

0. Preliminary Business

  • New WE1S participants, and participants transitioning to new status
    • Ray Stedding, CSUN graduate (to be WE1S Alumni Research Associate)
    • Avery Martin, UCSB (working in winter and spring quarters with WE1S as English 199RA student)
    • Jessica Gang, UCSB (working this quarter with WE1S as English 199RA student)
    • Cindy Kang, UCSB (working this quarter with WE1S as  MAT 597 Directed Research student)
  • New RAs to be introduced in future from U. Miami 

 

 

 

1. Project Roadmap for Rest of Academic Year 2018-2019

  1. Ongoing tasks with end-of-academic year deadlines
    1. Standing teams (collection, interpretation, and methods development work)
      1. Primary Corpus Collection & Analysis Team (led by Lindsay Thomas)
      2. Students and the Humanities Team (led by Abigail Droge)
      3. Diverse Populations Team (led by Giorgina Paiella)
      4. Interpretation Lab Team (led by Dan Baciu)
        1. Interpretation Protocol development
        2. Data Visualization 
        3. Word Embedding
    2. Collection of materials (Possible all-hands meeting in Feb. or Mar. to set goals for collection by end of year?)
      1. "Complete" U.S. collection by June?
        1. Current collection tally 
      2. "Random" corpus collection at U.M.
      3. Subcorpora:
        1. student-related material
        2. social media material
        3. "diverse populations" material
        4. Spanish-language corpus 
  2. Development of Interpretation Protocol 
    1. January: Preparatory tasks to improve topic modeling interpretation (see below for details)
    2. February and March: next series of topic model interpretation workshops.
    3. April, May, and June: "Real" interpretation of our corpus (including subcorpora) to address one to three specific research question--e.g.,
      • What is the similarity/difference between mainstream news media and student discourse on the humanities? (Or a comparison between our main corpus and any subcorpus)
      • How do the humanities and sciences compare in news media?
      • How does the "cosmic background radiation" of the humanities compare in the U.S. West vs. East vs. Midwest, vs. South?
      • How does our corpus compare to a "random" corpus? 
  3. Goal by end of academic year: position ourselves to address further rersearch questions and embark on analytical and reporting work on our corpus.
  4. Summer research camp 2019 (July 1-Aug 2) 

 

 

2. January: Preparatory tasks to improve topic modeling interpretation

  • Task 1. Corpus Selection Tasks (Develop methods of selecting, sampling, and normalizing corpus materials for modeling), led by Dan Baciu, probably in collaboration with Scott Kleinman as he rolls out the Workspace Management System -- Task Log

  • Task 2. Create Topic Models at Standard Set of Granularities (Task: Add code in Jupyter notebooks (Python workflow) to allow for optional creation of models at different levels of granularity), led by Jeremy -- Task Log

  • Tasks 4-7 Model Optimization Tasks -- Combined Task Log 

    • 4. Optimize models using Mallet diagnostics, led by Dan Baciu
    • 5. Optimize choosing best number of topics for a model, led by Dan Baciu
    • 6. Optimize with help of word embedding, team: Fabian Offert & Teddy Roland
    • 7. Human inspection of a model, led by Abigail Droge (draft protocol created as of 1/10/2010)

      (perhaps participants in this task could meet in January?)
  • Task 8: Stopwords & Consolidations, to be contributed to be all hands -- Page for gather suggestions for stopwords and consolidations

  • Tasks 9-10: Auto-Label Topics and Identify “Interesting” Features of a Topic Model, Team: Sihwa Park, Cindy Kang, Tyler Shoemaker -- Task Log

  • Task 11: Human Topic Modeling, led by Abigail Droge -- Task Log

  • Task 12. Macroanalysis of Topic Models (develop a method for this to add to our interpretation protocol), led by Dan Baciu and interpretation team -- Task Log

 

 

 

Planning for Future Meetings

  • Standing teams & other task teams to meet and work independently during January
  • Next all-hands meeting likely beginning in February (e.g., the next series of interpretation workshops) 

 

 

 

Optional: Breakout meetings

 

 

 

 

 

 

Comments (0)

You don't have permission to comment on this page.